Journal of the Japan Institute of Metals and Materials
Online ISSN : 1880-6880
Print ISSN : 0021-4876
ISSN-L : 0021-4876
Volume 57 , Issue 5
Showing 1-18 articles out of 18 articles from the selected issue
  • Hiromi Miura, Takashi Yoshida, Taku Sakai, Masaharu Kato, Tsutomu Mori
    1993 Volume 57 Issue 5 Pages 479-485
    Published: 1993
    Released: April 04, 2008
    JOURNALS FREE ACCESS
    Cu bicrystals with various [110] symmetric tilt boundaries doped with Bi were tensile tested. Their fracture behavior depended sensitively on the misorientation angle at the boundaries, Bi doping time and aging temperature. A strong correlation between the fracture behavior and grain-boundary energy before the Bi doping has been found: the higher the grain-boundary energy, the easier becomes the brittle fracture at grain boundaries. As the Bi doping time increased, grain-boundary brittle fracture occurred more easily. These phenomena could be explained resonably by considering the boundary-dependent amount of segregation of Bi atoms.
    Download PDF (1039K)
  • Naoya Fujita, Yuko Hanatate
    1993 Volume 57 Issue 5 Pages 486-491
    Published: 1993
    Released: April 04, 2008
    JOURNALS FREE ACCESS
    A Cu-Zn-Al alloy showing spontaneous shape changes at high temperatures was aged as the pre-deformed state, and the effects of the heat treatment, i.e., constrained aging, on the reversible shape memory effect, the inverse shape memory effect and the high temperature shape change, have been investigated as a function of deformation temperature, aging temperature and aging time.
    The results are summarized as follows; (1) When the aging temperature is lower than Af, both the amounts of the inverse shape memory effect and the high temperature shape change become nearly 0. (2) When the temperature is higher than Af, both of them increase with increasing temperature, and they show maximum values in the temperature range higher than 420 K. However, the amount of the inverse shape memory effect decrease with bainite transformation. (3) Under the conditions where a specimen shows the reversible shape memory effect, it shows the inverse shape memory effect and the high temperature shape change. (4) There is a good correlation between the amounts of the inverse shape memory effect and the high temperature shape change.
    Download PDF (679K)
  • Akira Takeuchi, Toshiyuki Koyama, Toru Miyazaki
    1993 Volume 57 Issue 5 Pages 492-500
    Published: 1993
    Released: April 04, 2008
    JOURNALS FREE ACCESS
    Computer simulations of phase decomposition are performed on the basis of the Cahn-Hilliard’s two-dimensional non-linear diffusion equation for Al-Zn and Fe-Mo binary alloys where the mottled and the modulated structures are formed during ageing. In the calculation, the modified regular solution approximation was adopted, i.e. the composition and temperature dependencies of the interaction parameter Ω was taken into account and the anisotropic elastic strain energy was also considered.
    The phase decomposition is successfully computed for Al-Zn and Fe-Mo alloys. The morphologies of the calculated microstructures are in good agreement with those of the experimentally obtained microstructures for both alloys. Moreover, the values of the mean wavelength of the calculated microstructures agree well with those of the experimentally obtained ones.
    Download PDF (1099K)
  • Hideki Araki, Yoritoshi Minamino, Toshimi Yamane, Shigeoki Saji, Satos ...
    1993 Volume 57 Issue 5 Pages 501-508
    Published: 1993
    Released: April 04, 2008
    JOURNALS FREE ACCESS
    The diffusion coefficient of W in β-Ti has been determined at temperatures from 1323 to 1673 K under pressures from 90 kPa to 2.8 GPa using the diffusion couple method. The values for the activation volume, ΔV, calculated from the pressure dependence of the diffusion coefficients at various temperatures, are (0.28∼0.41) V0, where V0 is the molar volume of Ti at room temperature and the atmospheric pressure. These values are considerably smaller than those measured for the bcc systems characterized by the “simple” monovacancy diffusion mechanism. We applied the model for phonon controlled diffusion to diffusion in β-Ti under high pressure and evaluated ΔV for self- and W diffusion in β-Ti as functions of T and ∂T0⁄∂P, where T is the temperature, P the pressure, and T0 the hypothetical temperature at which the bcc lattice becomes instable for atomic displacements in the ⟨111⟩ direction. The ΔV estimated for ∂T0⁄∂P=11 K/GPa is in agreement with both experimental results for W diffusion in β-Ti in the present work and for self-diffusion in β-Ti reported by Jeffery. This good agreement suggests that diffusion in β-Ti occurs with the mechanism of phonon-assisted diffusion jumps via monovacancies, which is characterized by a positive temperature-dependent migration enthalpy and a positive temperature-dependent migration volume.
    Download PDF (695K)
  • Hisashi Kuwano, Shinobu Ohkido, Yuichi Ishikawa, Toshihiko Yoshimura, ...
    1993 Volume 57 Issue 5 Pages 509-514
    Published: 1993
    Released: April 04, 2008
    JOURNALS FREE ACCESS
    A stainless steel SUS329J2L which has generally been in commercial use in the form of a duplex phase was heat-treated to ensure a single ferrite phase. The ferritic steel was aged at 723 K for times up to 10.8 Ms (3000 h) to clearify the mechanism of the phase decompositon of ferrite. Precipitation of Cr-rich α phase was investigated by difference Mössbauer spectra and TEM observation. A time evolution of the compositional fluctuation was evaluated on the basis of the probabilities of the hyperfine magnetic field distribution computed from the Mössbauer spectra. The phase decomposition proceeded phenomenologically in three stages to yield finally a mixture of Fe-rich α and Cr-rich α phases. In the initial stage of the decomposition, the formation of α phase and spinodal-like compositional fluctuation occurred simultaneously. In the following stage, the precipitation of α phase proceeded with increasing aging time, whereas the compositional fluctuations grew slower. The final stage was characterized by coalescence of α phases.
    Download PDF (866K)
  • Kaoru Tada, Yoshimi Hisatsune, Masao Yamamoto
    1993 Volume 57 Issue 5 Pages 515-520
    Published: 1993
    Released: April 04, 2008
    JOURNALS FREE ACCESS
    Phase change on the surface of the Fe-Cr-Mn alloy through the vaporization of Mn, which has a high vapor pressure, was studied in order to develop a new surface modification method utilizing the vapor pressure differences among the alloy elements. Fe-(10-30 mass%)Cr-(10-30 mass%)Mn alloys were heated at 1373 K, and the phases of the samples were examined before and after the heat treatment. The microstructures of the cross section of the samples were observed, and the concentration depth profiles of the alloy elements were measured after the heat treatment.
    The surface concentrations of the heat treated samples depended strongly on the atmospheres during heat treatment. The Mn concentration on the sample surfaces heat treated in a closed tube were 7-15% and those heat treated under a constant pressure using a vacuum pump were 1-2%. The gamma phase alloys, containing more than 20%Mn, changed to the alpha phase and carbide due to such shift in the surface concentration. The sigma phase was also formed on the surface of the Fe-20%Mn-20%Cr alloy. Surface layers with different microstructures from the base alloy were observed. These surface layers were 10-50 μm thick and corresponded to the newly formed phases. On the other hand, no phase change was observed for the alpha phase alloys, containing less than 15%Mn.
    It seems possible that the sigma phase layer is formed singularly on the surface of the Fe-Cr-Mn alloy through controlling the alloy compositions and the heat treatment conditions for Mn vaporization.
    Download PDF (995K)
  • Hiroyuki Fukuyama, Toshiharu Fujisawa, Chikabumi Yamauchi
    1993 Volume 57 Issue 5 Pages 521-526
    Published: 1993
    Released: April 04, 2008
    JOURNALS FREE ACCESS
    The activity of SbO2.5 in NaO0.5-CO2-SbOm slag was calculated by integrating the Gibbs-Duhem relation from the activity of NaO0.5 determined in the previous experimental study by the authors, over the composition range of 1>N>0.7 {=nNaO0.5⁄(nNaO0.5+nSbOm), n: the number of moles} under the partial pressure of CO2 of 0.001∼0.08 MPa at the temperature of 1423 and 1523 K. Since it was found that Sb in the slag changed its valence depending on the experimental conditions, the redox equilibria of Sb in the slag were taken into consideration for the selection of the suitable starting point of the integration.
    The activity of Na3SbO4 was also discussed and the standard free energy change for the formaiton of Na3SbO4(l) was determined as follows:
    (This article is not displayable. Please see full text pdf.)
    It was possible to calculate the content of Sb in molten copper equilibrated with the slag based on the obtained results. Thus, combined with the solubility data on CO2 in the slag reported in the previous study, the equilibrium distribution ratio of Sb{LSb=(mass% of Sb in slag)/[mass% of Sb in molten copper]} was also calculated over the wide slag composition range as a function of partial pressures of CO2 and O2.
    Download PDF (471K)
  • Norihito Ikemiya, Jun Yoshitomi, Shigeta Hara, Kazumi Ogino
    1993 Volume 57 Issue 5 Pages 527-532
    Published: 1993
    Released: April 04, 2008
    JOURNALS FREE ACCESS
    Surface tensions and densities of melts in the TiO2-BaO and TiO2-Na2O binary systems were measured by the maximum bubble pressure method accurately. The results obtained were as follows:
    (1) Surface tensions of melts in the TiO2-BaO binary system are in agreement with calculated values from Guggenheim’s equation for ideal solution near the TiO2-50 mol%BaO composition, but for melts in the high TiO2 concentration region they deviate positively from calculated values for ideal solution.
    While for melts in the TiO2-Na2O binary system they deviate negatively from the calculated values for ideal solution.
    (2) Molar volumes of these two binary systems deviate positively from the additive rule on volume.
    (3) Molar volumes of molten MXO·TiO2(X = 1 for M: Ti2+, Ba2+ and X = 2 for M: Na+) near the melting point increase linearly as a function of Xrc3(rc: the ionic radius of the cation M).
    (4) Surface tensions of molten MXO·TiO2 near the melting point increase as the ionic potentials of the cation M increase.
    Download PDF (416K)
  • Hidehiko Kamide, Yuichi Tanaka
    1993 Volume 57 Issue 5 Pages 533-537
    Published: 1993
    Released: April 04, 2008
    JOURNALS FREE ACCESS
    Cyclic hot corrosion behaviour of the NiAl intermetallic compound which was changed Al and Si contents were examined by using a fused 75 mass%Na2SO4-25 mass%NaCl mixture. Specimens were immersed in fused salt at 1173 K for 10.8 ks and then surface scales were removed. This process was repeated 5 cycles.
    Hot corrosion resistance was increased with increasing Si content. This behaviour was especially recognized in high Al concentration specimens. On the other hand, microscopic observation showed that the grain-boundary attack was observed in all specimens and the depth of this attack was prevented in Si-containing NiAl alloys.
    X-ray distribution profiles for Ni and Al showed that a compact Al2O3 layer was formed on the surface of Si-containing NiAl alloy and the thickness of the low Al concentration zone under the Al2O3 layer was decreased with the formation of the compact Al2O3 layer. The X-ray distribution image showed that an amorphous Si oxide was formed outside the compact Al2O3 layer.
    It is thought that the formation of the compact Al2O3 layer which is accelerated with the formation of the amorphous Si oxide improves the hot corrosion resistance during corrosion tests and the decrease in the thickness of the low Al concentration zone which is prevented with the formation of the compact Al2O3 layer improve the hot corrosion resistance after the scale removal during cyclic hot corrosion tests.
    Download PDF (571K)
  • Yoshinori Wakamatsu, Makoto Izumida, Toshitada Shimozaki, Masami Onish ...
    1993 Volume 57 Issue 5 Pages 538-543
    Published: 1993
    Released: April 04, 2008
    JOURNALS FREE ACCESS
    In order to obtain fundamental information on TiC coating on carbon steel using Pb-Ti melt, solid state reaction of S15C, S25C, S35C, S45C, S55C and SK5 specimens in molten Pb containing 0.1 to 5 mass%Ti at 1073 to 1373 K has been investigated. A dense and uniform TiC phase layer having Vickers hardness of 3000∼3500 was obtained on the surfaces of the specimens, although the TiC layer formed in molten Pb containing more than 1 mass%Ti was covered with a porous Ti2PbC phase layer having Vickers hardness of about 100. Thickness of the TiC layer increased with increasing Ti content of molten Pb, C content of the specimens, annealing time and temperature, while any definite relation could not be found between thickness of the Ti2PbC layer and annealing conditions. It is concluded that the TiC layer grows at the interface between TiC layer and molten Pb or Ti2PbC layer and that the growth rate is controlled by diffusion process of C atoms in the TiC layer.
    Download PDF (682K)
  • Kazuo Kasahara, Atsushi Takei
    1993 Volume 57 Issue 5 Pages 544-548
    Published: 1993
    Released: April 04, 2008
    JOURNALS FREE ACCESS
    The oxidation of a Ti-50 mol%Al alloy has been investigated at temperatures from 873 to 1073 K in air using cyclic (18 ks×10 cycles) or continuous (180 ks) oxidation tests for as-cast and homogenized specimens. The microstructure of the as-cast specimen has a lamellar structure consisting of the α2 and γ phases, while the homogenized specimen consists of the γ single phase.
    The mass gains of both specimens after cyclic or continuous oxidation at 873 K were very small. At 973 K, the mass gain of the homogenized specimen was small for the cyclic oxidation test, compared with that of the as-cast specimen containing a large amount of the α2 phase. Thus it was concluded that the oxidation resistance of TiAl at 973 K was apparently influenced by the microstructure. At 1073 K, the mass gain of the homogenized specimen apparently increased during cyclic oxidation with fracture of the oxide scale on the specimen surface by thermal stress, because the specimen was covered with a large amount of the oxide scale of Al2O3 which exhibited poor adhesion to the matrix. The mass gain of the as-cast specimen was smaller than that of the homogenized specimen in cyclic oxidation. This indicates that the adhesion of the oxide scale is affected by the microstructure. However, the mass gain of both specimens after continuous oxidation was small compared with that of the cyclic oxidation. It was considered that the adhesion of the oxide scale is not affected by the thermal stress during the continuous oxidation.
    Download PDF (777K)
  • Isamu Yuitoo, Masahiro Kitada
    1993 Volume 57 Issue 5 Pages 549-554
    Published: 1993
    Released: April 04, 2008
    JOURNALS FREE ACCESS
    The Effect of surface oxidation at the interface between Ni-Fe and Fe-Mn alloy thin layers on the exchange coupling of an Fe-Mn/Ni-Fe bilayer thin film was investigated. The Ni82-Fe18 alloy film was deposited by an electron beam system in a vacuum of 10−4 Pa. The deposited film was exposed to air for 72 h in order to oxidize the surface of the Ni-Fe layer. The surface of the Ni-Fe layer was etched by ion-milling to clean the oxidized surface. The Fe-Mn alloy film was deposited by ion beam sputtering, using a Fe50Mn50 target. The exchange coupling field was obtained from the shift in the B-H loop. The following can be concluded from the results. Exchange coupling does not occur, when the Ni-Fe surface is oxidized, due to the growth of paramagnetic α-(Fe-Mn). However, if the Fe-Mn film is deposited immediately after etching the Ni-Fe surface, exchange coupling is observed. This is due to the epitaxial growth of antiferromagnetic γ-(Fe-Mn) on the clean surface. The exchange coupling field is comparable to the value for the Ni-Fe/Fe-Mn multilayer deposited continuously. The exchange coupling field becomes smaller when the Fe-Mn layer is thinner than 10 nm, because Mn is preferentially oxidized.
    Download PDF (638K)
  • Hiroyuki Takeda, Mikako Tsujimoto, Kiyoshi Nogi, Takao Sakata, Kazumi ...
    1993 Volume 57 Issue 5 Pages 555-560
    Published: 1993
    Released: April 04, 2008
    JOURNALS FREE ACCESS
    The (100), (110) and (111) planes of MgO were observed at an atomic scale using an atomic force microscope. It was identified that the surface structures were different from the bulk ones and that the surface defects existed on each surface. It can be suggested that the surface free energy and the polarization play an important role in the alteration of the surface structures. It can be considered that all the bright spots in the AFM images correspond to the positions of the oxygen atom in the case of the MgO surface.
    Download PDF (480K)
  • Toshio Shimoo, Yasuhiro Kobayashi, Kiyohito Okamura
    1993 Volume 57 Issue 5 Pages 561-567
    Published: 1993
    Released: April 04, 2008
    JOURNALS FREE ACCESS
    The kinetics and the mechanism of the reaction of Si3N4 with Cr have been investigated. Using powdered Si3N4-Cr mixture, the reaction rate was determined by means of a thermo-balance in an Ar gas stream at temperatures from 1273 to 1573 K. The reaction products were examined by X-ray diffraction.
    The reaction of Si3N4 with Cr took place after heating to about 1250 K. Cr2N and Cr3Si were produced below 1373 K. Above 1473 K, Cr2N disappeared and Cr3Si changed to CrSi via Cr5Si3. The initial rate was described by the linear rate law. The activation energy was 260 kJ/mol. A chemical process may be considered as the rate-determining step. When the Cr particles were covered with the reaction product layer, the kinetics obeyed the parabolic rate law. The activation energy was 365 kJ/mol. The reaction rate is probably controlled by the diffusion through the reaction layer.
    Download PDF (530K)
  • Akira Sato, Yoshiaki Osawa, Goro Aragane
    1993 Volume 57 Issue 5 Pages 568-573
    Published: 1993
    Released: April 04, 2008
    JOURNALS FREE ACCESS
    A production method of aluminum alloy profile rods by the moldless upward continuous casting process using formers was studied. Effects of the composition and the superheat of molten alloy, the withdrawal speed, and the size of the nozzle made in the former, on the ratio of strand formation (the ratio of the size of the cast strand to the size of the former nozzle, %), the sectional configuration, the macrostructure, the microstructure, and the surface conditions of the cast rod obtained were examined at a constant intensity of the strand cooling. Aluminum alloy profile rods having similar configurations to those of former nozzles and longitudinal tapers can be produced. The former nozzle determines the configuration of the molten alloy pillar formed on withdrawing upwardly. The square or asterisk cast rod is composed of columnar grains grown completely in the longitudinal direction. The ratio of strand formation is governed by the combination of the composition and the superheat of the molten alloy, the size of the former nozzle, and the withdrawal speed. The alloy profile rod having no longitudinal taper can be produced most easily under the condition where the ratio of strand formation is around 85∼95%.
    Download PDF (1234K)
  • Michiko Yoshihara, Naochika Imamura, Equo Kobayashi, Kenji Miura, Yosh ...
    1993 Volume 57 Issue 5 Pages 574-581
    Published: 1993
    Released: April 04, 2008
    JOURNALS FREE ACCESS
    The cyclic oxidation resistance in air was investigated mainly at 900 and 950°C for stoichiometric TiAl intermetallics alloyed with up to 5 at%Nb and Ti-Al-Nb ternary alloys with 50 at%Al and 0-20 at%Nb.
    The mass gains of most of the Nb-added alloys due to oxidation are significantly less than those of the binary TiAl, indicating that the oxidation resistance is remarkably improved by Nb addition. The alloys containing more than 15 at%Nb show a relatively larger mass gain owing to the TiO2·Nb2O5 formation. Oxide scales formed during oxidation are similar in morphology to those formed on TiAl, become thinner as the Nb content increased and the Al2O3 layer in the scales becomes denser and continuous. The improvement of oxidation resistance is attributed to both the suppression of TiO2 growth and the promotion of protective Al2O3 layer formation. Substitution of Nb+5 into the TiO2 is expected to decrease the oxygen vacancy concentration and to reduce the growth rate of TiO2. The content of Nb which is just enough to eliminate oxygen vacancies is estimated to be 2 at%, while Ti-Al-Nb containing about 10 at%Nb shows the best oxidation resistance. Hence, it is suggested that Nb also affects the diffusion behaviour of Al so as to enhance the formation of Al2O3. The alloys containing a few percent Nb, when heat-treated under a low partial pressure oxygen atmosphere, show better oxidation resistance than binary TiAl with the same heat treatment. It is presumed that an Al2O3 layer formed on the alloys by the treatment is thicker than that formed on TiAl, which supports the above suggestion. The heat treatment, however, is not effective in improving the oxidation resistance of the alloys with more than 10 at%Nb. It is probably because that the β phase appeares in Al-depleted region which forms as an Al2O3 grows, thus resulting in the formation of thick internal oxidation layer.
    Download PDF (1027K)
  • Yi Tan, Tetsumori Shinoda, Yoshinao Mishima, Tomoo Suzuki
    1993 Volume 57 Issue 5 Pages 582-590
    Published: 1993
    Released: April 04, 2008
    JOURNALS FREE ACCESS
    The effects of deviation from stoichiometry and addition of boron have been investigated on the creep behaviour of NiAl. It is found that the creep behaviour of near stoichiometric NiAl alloys is of the class II type with a stress exponent, n, being 4.5, while that of off-stoichiometric alloys of the class I type with n being 3.5. The activation energy for creep, QC, decreases with increasing Al concentration at the Ni-rich side of offstoichiometry but vice versa on the Al-rich side of offstoichiometry. At near stoichiometry, the values for the QC are off the relation being found at offstoichiometry and are almost constant within 1 at% on both sides of stoichiometry. The QC seems to correspond to the value of the activation energy for self diffusion of Ni at near stoichiometry and to that for interdiffusion at offstoichiometry.
    The effect of boron addition is found to increase the creep resistance of NiAl alloys at Ni-rich side of offstoichiometry without an appreciable change in the stress exponent. The reason for the enhancement of creep resistance should be due to the solution hardening by boron. At Al-rich side of offstoichiometry, no effect on the creep resistance is found by the boron addition.
    Download PDF (776K)
  • Yasushi Fukuzawa, Naotake Mohri, Masayoshi Tateno, Yo Kojima
    1993 Volume 57 Issue 5 Pages 591-595
    Published: 1993
    Released: April 04, 2008
    JOURNALS FREE ACCESS
    Possibility of decreasing the residual stress on the metal-ceramics interface was investigated with a TiB2-Ni system. The ceramics surface was deformed by an electrical discharge machining with a metal electrode. In this study, outer and/or inner shapes effects were investigated with a cylindrical joining material. Six joining shapes were produced by a direct diffusion bonding method under a high vacuum and high temperature condition.
    The following results are obtained:
    In the changing of outer shapes, the highest bonding strength is obtained at the suitable joining temperature region and the interface slop. On the suitable joining temperature region, the fracture behavior transfers from interface fracture to ceramics side failure. Also the bonding strength and fracture behaviors are affected by the inner and the outer interface shapes. From the abovementioned results, the bonding strength will be controlled by the interface surface shapes and the bonding temperature.
    Download PDF (435K)
feedback
Top